Search results

The study aims to develop a multilayer high-effective ensemble of ensembles predictive model (stacking ensemble) using several hyperparameter optimized ensemble machine learning (ML) methods (bagging and boosting ensembles) trained with high-volume data points retrieved from Internet of Things (IoT) emission sensors, time-corresponding meteorology and traffic data.

For a start, the study experimented big data hypothesis theory by developing sample ensemble predictive models on different data sample sizes and compared their results. Second, it developed a standalone model and several bagging and boosting ensemble models and compared their results. Finally, it used the best performing bagging and boosting predictive models as input estimators to develop a novel multilayer high-effective stacking ensemble predictive model.

Results proved data size to be one of the main determinants to ensemble ML predictive power. Second, it proved that, as compared to using a single algorithm, the cumulative result from ensemble ML algorithms is usually always better in terms of predicted accuracy. Finally, it proved stacking ensemble to be a better model for predicting PM_2.5 concentration level than bagging and boosting ensemble models.

A limitation of this study is the trade-off between performance of this novel model and the computational time required to train it. Whether this gap can be closed remains an open research question. As a result, future research should attempt to close this gap. Also, future studies can integrate this novel model to a personal air quality messaging system to inform public of pollution levels and improve public access to air quality forecast.

The outcome of this study will aid the public to proactively identify highly polluted areas thus potentially reducing pollution-associated/ triggered COVID-19 (and other lung diseases) deaths/ complications/ transmission by encouraging avoidance behavior and support informed decision to lock down by government bodies when integrated into an air pollution monitoring system

This study fills a gap in literature by providing a justification for selecting appropriate ensemble ML algorithms for PM_2.5 concentration level predictive modeling. Second, it contributes to the big data hypothesis theory, which suggests that data size is one of the most important factors of ML predictive capability. Third, it supports the premise that when using ensemble ML algorithms, the cumulative output is usually always better in terms of predicted accuracy than using a single algorithm. Finally developing a novel multilayer high-performant hyperparameter optimized ensemble of ensembles predictive model that can accurately predict PM_2.5 concentration levels with improved model interpretability and enhanced generalizability, as well as the provision of a novel databank of historic pollution data from IoT emission sensors that can be purchased for research, consultancy and policymaking.

This study aims to develop a comprehensive conceptual framework that serves as a foundation for identifying most critical delay risk drivers for Building Information Modelling (BIM)-based construction projects.

A systematic review was conducted using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) to identify key delay risk drivers in BIM-based construction projects that have significant impact on the performance of delay risk predictive modelling techniques.

The results show that contractor related driver and external related driver are the most important delay driver categories to be considered when developing delay risk predictive models for BIM-based construction projects.

This study contributes to the body of knowledge by filling the gap in lack of a conceptual framework for selecting key delay risk drivers for BIM-based construction projects, which has hampered scientific progress toward development of extremely effective delay risk predictive models for BIM-based construction projects. Furthermore, this study's analyses further confirmed a positive effect of BIM on construction project delay.

This paper seeks to assess the performance levels of BA-GS-LSSVM compared to popular standalone algorithms used to build NO₂ prediction models. The purpose of this paper is to pre-process a relatively large data of NO₂ from Internet of Thing (IoT) sensors with time-corresponding weather and traffic data and to use the data to develop NO₂ prediction models using BA-GS-LSSVM and popular standalone algorithms to allow for a fair comparison.

This research installed and used data from 14 IoT emission sensors to develop machine learning predictive models for NO₂ pollution concentration. The authors used big data analytics infrastructure to retrieve the large volume of data collected in tens of seconds for over 5 months. Weather data from the UK meteorology department and traffic data from the department for transport were collected and merged for the corresponding time and location where the pollution sensors exist.

The results show that the hybrid BA-GS-LSSVM outperforms all other standalone machine learning predictive Model for NO₂ pollution.

This paper's hybrid model provides a basis for giving an informed decision on the NO₂ pollutant avoidance system.

This research installed and used data from 14 IoT emission sensors to develop machine learning predictive models for NO₂ pollution concentration.

This paper aims to establish the most underlying factors causing construction projects delay from the most applicable.

The paper conducted survey of experts using systematic review of vast body of literature which revealed 23 common factors affecting construction delay. Consequently, this study carried out reliability analysis, ranking using the significance index measurement of delay parameters (SIDP), correlation analysis and factor analysis. From the result of factor analysis, this study grouped a specific underlying factor into three of the six applicable factors that correlated strongly with construction project delay.

The paper finds all factors from the reliability test to be consistent. It suggests project quality control, project schedule/program of work, contractors’ financial difficulties, political influence, site conditions and price fluctuation to be the six most applicable factors for construction project delay, which are in the top 25% according to the SIDP score and at the same time are strongly associated with construction project delay.

This paper is recommending that prospective research should use a qualitative and inductive approach to investigate whether any new, not previously identified, underlying factors that impact construction projects delay can be discovered as it followed an inductive research approach.

The paper includes implications for the policymakers in the construction industry in Nigeria to focus on measuring the key suppliers’ delivery performance as late delivery of materials by supplier can result in rescheduling of work activities and extra time or waiting time for construction workers as well as for the management team at site. Also, construction stakeholders in Nigeria are encouraged to leverage the amount of data produced from backlog of project schedules, as-built drawings and models, computer-aided designs (CAD), costs, invoices and employee details, among many others through the aid of state-of-the-art data driven technologies such as artificial intelligence or machine learning to make key business decisions that will help drive further profitability. Furthermore, this study suggests that these stakeholders use climatological data that can be obtained from weather observations to minimize impact of bad weather during construction.

This paper establishes the three underlying factors (late delivery of materials by supplier, poor decision-making and Inclement or bad weather) causing construction projects delay from the most applicable.

Several factors influence OSC adoption, but extant literature did not articulate the dominant barriers or drivers influencing adoption. Therefore, this research has not only ventured into analyzing the core influencing factors but has also employed one of the best-known predictive means, Machine Learning, to identify the most influencing OSC adoption factors.

The research approach is deductive in nature, focusing on finding out the most critical factors through literature review and reinforcing — the factors through a 5- point Likert scale survey questionnaire. The responses received were tested for reliability before being run through Machine Learning algorithms to determine the most influencing OSC factors within the Nigerian Construction Industry (NCI).

The research outcome identifies seven (7) best-performing algorithms for predicting OSC adoption: Decision Tree, Random Forest, K-Nearest Neighbour, Extra-Trees, AdaBoost, Support Vector Machine and Artificial Neural Network. It also reported finance, awareness, use of Building Information Modeling (BIM) and belief in OSC as the main influencing factors.

Data were primarily collected among the NCI professionals/workers and the whole exercise was Nigeria region-based. The research outcome, however, provides a foundation for OSC adoption potential within Nigeria, Africa and beyond.

The research concluded that with detailed attention paid to the identified factors, OSC usage could find its footing in Nigeria and, consequently, Africa. The models can also serve as a template for other regions where OSC adoption is being considered.

The research establishes the most effective algorithms for the prediction of OSC adoption possibilities as well as critical influencing factors to successfully adopting OSC within the NCI as a means to surmount its housing shortage.

The study aims to analyse the role of digitisation in accounting in enhancing the overall effectiveness of accounting functions. To achieve this, the study provides empirical evidence from the stakeholder’s perspective of digitisation of accounting, auditing, reporting and regulatory compliance procedures.

The study has applied a quantitative approach to identify the thoughts of auditors, accountants and academicians on the impact of digitalised accounting applications on accounting functions. The data was collected by administering an empirical study and a sample of 482 professionals from the accounting, auditing and academic sectors. To analyse and interpret data descriptive statistics, structured equation modelling and mediation analysis has been used.

The finding of the study signifies the relevance of digitalised accounting applications in accounting functions and reveals that there is a significant impact of digitalisation on accounting, auditing, reporting and regulatory compliance aspects of accounting functions. The outcome of the study explores that a digitalised accounting system reduces possible errors and improves the accuracy and transparency of the system.

The study highlighted the importance of developing new methods and techniques that can be used in practice. This indirectly advocates the inclusion of such concepts in accounting curricula to emphasise the need to understand the challenges and opportunities created by digitisation. Furthermore, the study will become a motivation to scholars who intend to explore different areas through which new technologies can be adopted to transform traditional accounting systems.

The contributions of the current study have implications that the adoption of digitised accounting enhances economic efficiency through a reduction in accounting costs, and enhanced accuracy that leads to the elimination of penalties and litigations for non-compliance with regulatory authorities. This indirectly impacts positively on the financial health of the business organisations and economies at large. This implication becomes greater evidential support to the organisations which are yet to plan the adoption and implementation of digital tools in their organisation for accounting functions.

Digitalisation is a relevant part of the accounting function to improve efficiency and accuracy. Since accounting and auditing practitioners struggle to control the accuracy and efficiency of transactions. Furthermore, the outcome of the study assists organisations in gaining real-time access to financial data, transforms workflows and empowers management to make timely informed sound decisions, optimise resource allocation, efficient regulatory compliance and so on.

The purpose of this paper is to examine the impact of microfinance programs sponsored by Sudanese microfinance institutions (SMFIs) on monetary poverty reduction in Sudan where poverty is widely spread.

The study adopted the control group approach, where income and expenditure are taken as welfare indicators. The updated World Bank’s international poverty line of 1.90 per person per day was adopted to separate the poor from non-poor. The data were collected by the means of a questionnaire distributed to a random sample of beneficiaries in the institution under study. The study adapted the Foster, Greer and Thorbecke (FGT) model to evaluate the role of microfinance programs in poverty reduction. Furthermore, to gain more insight into the impact of the program, a preliminary analysis was conducted using the independent-samples t-test to examine the difference in the welfare indicators for the sample of the control group and treatment group as well as that of the small loan group and micro-loan group.

The findings show that the microfinance program provided by SMFIs has reduced the monetary poverty among the participants. The results also reveal that beneficiaries who had received a larger volume of loan were noted lesser poverty than those who had received very small loan size. Moreover, the results demonstrate that poverty indices based on expenditure as a welfare indicator are far lower than those based on income for both groups.

This study contributes to the available literature by filling the gaps through including income and expenditure as monetary variables, which included separately in previous studies adopted the FGT model in the area of microfinance, in addition to exploring the role of loan size in the effect of microfinance on poverty reduction.

This paper aims to appraise the roles of whistleblowing policy as a tool for combating corruption in Nigeria. Methodologically, it examines how the policy could be strengthened to effectively address the challenges of corruption in Nigeria.

This paper is essentially a desk research with reliance on the secondary source of data. Relevant materials were collected in an eclectic manner from official documents, statutes and other published outlets such as books, journal publications, online articles, news reports and newspaper articles. Its scope is limited to issue and content analysis relating to the use of whistleblowing policy as a tool to combat corruption.

The paper finds that whistleblowing policy is an effective anti-corruption instrument that has facilitated discovery and recovery of looted public resources and prosecution of culprits in Nigeria.

This paper demonstrates how whistleblowing as an anti-corruption mechanism could be strengthened in Nigeria when the legislator finally passed the Whistleblower Protection Bill into law.

The purpose of this paper therefore is to identify and examine major issue-areas in law, prominent among which are the Plea-Bargain and S308 Immunity Clause, and how they impact the process of effectively combating corruption in Nigeria.

The paper uses documentary sources and analytical method to examine the issues involved.

The identified issue-areas are inhibitors rather than facilitators.

The implication is that the government needs to change the existing laws to strengthen the fight against corruption.

This is to ensure that the war against corruption is strengthened and effective.

To ensure that offenders face the full weight of the law for their action.

This paper is the author's original work and all references are appropriately acknowledged.

This paper aims to assess if the Economic and Financial Crimes Commission (EFCC) has been effective in combating corruption in Nigeria from 2003-2012.

The paper adopted a documentary analytical approach.

The organization has not been effective in combating corruption in Nigeria.

The study is between 2003-2012.

There is a need to correct those identified inhibitors that undermined the Commission’s capacity, such as intrusive government interference, lack of autonomy, poor funding and weak laws, among others, to mitigate corruption.

Eliminating those identified constraints will remove the incentive to be corrupt, thereby curbing the desire to be corrupt.

This paper is an original assessment of the EFCC's effectiveness in combating corruption in Nigeria during the specified period.